Invertebrates are dominant species in primary tropical rainforests, where their abundance and diversity contributes to the functioning and resilience of these globally important ecosystems. However, more than one-third of tropical forests have been logged, with dramatic impacts on rainforest biodiversity that may disrupt key ecosystem processes. We find that the contribution of invertebrates to three ecosystem processes operating at three trophic levels (litter decomposition, seed predation and removal, and invertebrate predation) is reduced by up to one-half following logging. These changes are associated with decreased abundance of key functional groups of termites, ants, beetles and earthworms, and an increase in the abundance of small mammals, amphibians and insectivorous birds in logged relative to primary forest. Our results suggest that ecosystem processes themselves have considerable resilience to logging, but the consistent decline of invertebrate functional importance is indicative of a human-induced shift in how these ecological processes operate in tropical rainforests.

Forested tropical landscapes around the world are being extensively logged and converted to agriculture, with serious consequences for biodiversity and potentially ecosystem functioning. Here we investigate associations between habitat disturbance and functional diversity of ants and termites—two numerically dominant and functionally important taxa in tropical rain forests that perform key roles in predation, decomposition, nutrient cycling and seed dispersal. We compared ant and termite occurrence and composition within standardised volumes of soil and dead wood in old growth forest, logged forest and oil palm plantation in Sabah, Malaysian Borneo. Termites occurred substantially less frequently in converted habitats than in old growth forest, whereas ant occurrences were highest in logged forest and lowest in old growth forest. All termite feeding groups had low occurrence in disturbed habitats, with soil feeders occurring even less frequently than wood feeders. Ant functional groups showed more variable associations, with some opportunist and behaviourally dominant groups being more abundant in degraded habitats. The importance of ants and termites in tropical ecosystems and such differing patterns of assemblage variation suggest that ecosystem functioning may be significantly altered in converted habitats.

Functional diversity (FD) is an important component of biodiversity that quantifies the difference in functional traits between organisms. However, FD studies are often limited by the availability of trait data and FD indices are sensitive to data gaps. The distribution of species abundance and trait data, and its transformation, may further affect the accuracy of indices when data is incomplete. Using an existing approach, we simulated the effects of missing trait data by gradually removing data from a plant, an ant and a bird community dataset (12, 59, and 8 plots containing 62, 297 and 238 species respectively). We ranked plots by FD values calculated from full datasets and then from our increasingly incomplete datasets and compared the ranking between the original and virtually reduced datasets to assess the accuracy of FD indices when used on datasets with increasingly missing data. Finally, we tested the accuracy of FD indices with and without data transformation, and the effect of missing trait data per plot or per the whole pool of species. FD indices became less accurate as the amount of missing data increased, with the loss of accuracy depending on the index. But, where transformation improved the normality of the trait data, FD values from incomplete datasets were more accurate than before transformation. The distribution of data and its transformation are therefore as important as data completeness and can even mitigate the effect of missing data. Since the effect of missing trait values pool-wise or plot-wise depends on the data distribution, the method should be decided case by case. Data distribution and data transformation should be given more careful consideration when designing, analysing and interpreting FD studies, especially where trait data are missing. To this end, we provide the R package “traitor” to facilitate assessments of missing trait data.

There is a weak evidence base supporting the effective management of riparian ecosystems within tropical agriculture. Policies to protect riparian buffers—strips of non‐cultivated land alongside waterways—are vague and vary greatly between countries. From a rapid evidence appraisal, we find that riparian buffers are beneficial to hydrology, water quality, biodiversity and some ecosystem functions in tropical landscapes. However, effects on connectivity, carbon storage and emissions reduction remain understudied. Riparian functions are mediated by buffer width and habitat quality, but explicit threshold recommendations are rare. Policy implications. A one‐size fits all width criterion, commonly applied, will be insufficient to provide all riparian functions in all circumstances. Context‐specific guidelines for allocating, restoring and managing riparian buffers are necessary to minimise continued degradation of biodiversity and ecosystem functioning in tropical agriculture.

Summary: Dragonfly assemblages (Odonata: comprising damselflies, Zygoptera; and dragonflies, Anisoptera) in Southeast Asian rainforests are extremely diverse but increasingly threatened by habitat disturbance, including logging and conversion of forest to oil palm plantations. Land‐use change can affect dragonfly larval stages by altering within‐stream environmental conditions, and adults by loss of perches, shade and hunting habitat. However, the extent to which dragonflies are affected by land‐use change is not well known, and strategies for conservation are poorly developed. We surveyed dragonfly adults and larvae, forest quality and stream environmental conditions across 16 streams in Sabah, Malaysia. Habitat surrounding the streams included pristine forest, selectively logged forest, oil palm with forested riparian buffer strips and oil palm without buffers. Overall abundance and species richness of adult dragonflies stayed constant with habitat disturbance, but larval abundance and richness decreased with higher habitat disturbance, and larvae were largely absent from oil palm streams. There was also a clear shift in community composition of both adult and larval dragonflies. Anisoptera adults were more species rich and abundant, but Zygoptera adults were less species rich in more disturbed sites. The presence of riparian buffers in oil palm plantations offered some protection for forest‐associated dragonfly species, and streams with wider riparian buffers supported adult assemblages more similar to those found in logged forest. However, oil palm streams with riparian buffers still contained a depauperate larval assemblage compared to logged forest areas, and dragonfly assemblages in narrow riparian buffer streams were similar to those found in streams surrounded by continuous oil palm. Our results provide clear evidence of the effect of land‐use change on dragonflies. Conservation efforts to conserve forest communities should target the preservation of existing forest areas, but management within oil palm plantation landscapes to preserve riparian buffers can still have a marked beneficial effect on dragonfly communities.

Freshwaters provide valuable habitat and important ecosystem services but are threatened worldwide by habitat loss and degradation. In Southeast Asia, rainforest streams are particularly threatened by logging and conversion to oil palm, but we lack information on the impacts of this on freshwater environmental conditions, and the relative importance of catchment versus riparian‐scale disturbance. We studied 16 streams in Sabah, Borneo, including old‐growth forest, logged forest, and oil palm sites. We assessed forest quality in riparian zones and across the whole catchment and compared it with stream environmental conditions including water quality, structural complexity, and organic inputs. We found that streams with the highest riparian forest quality were nearly 4 °C cooler, over 20 cm deeper, had over 40% less sand, greater canopy cover, more stored leaf litter, and wider channels than oil palm streams with the lowest riparian forest quality. Other variables were significantly related to catchment‐scale forest quality, with streams in the highest quality forest catchments having 40% more bedrock and 20 times more dead wood, along with higher phosphorus, and lower nitrate‐N levels compared to streams with the lowest catchment‐scale forest quality. Although riparian buffer strips went some way to protecting waterways, they did not maintain fully forest‐like stream conditions. In addition, logged forest streams still showed signs of disturbance 10–15 years after selective logging. Our results suggest that maintenance and restoration of buffer strips can help to protect healthy freshwater ecosystems but logging practices and catchment‐scale forest management also need to be considered.